Earthquake prediction time domain analysis method for extracting precursor information based on earthquake wave decomposition

Abstract

The invention belongs to the technical field of earthquake prediction, and discloses an earthquake prediction time domain analysis method for extracting precursor information based on earthquake wavedecomposition. Dynamic modeling is carried out on propagation of earthquake waves in an earth crust medium by using an NLARI process, and parameter estimation is carried out on a dynamic model by using earthquake period whole-process data to obtain a dynamic parameter time sequence reflecting the whole process from beginning development, middle evolution to end of an earthquake; and propagation ofthe earthquake waves in the earth crust medium is modeled based on the NLARI process, initial change information of earth crust stress and rock friction force prior to strain is extracted through a technology of earthquake wave decomposition through model parameters, and earthquake prediction is carried out from two aspects of time and place. The invention provides the earthquake wave decomposition technology for obtaining the parameter time sequence reflecting earth crust dynamic changes in an earthquake period, and important information is provided for earthquake development and evolutionanalysis after the earthquake and aftershock prediction.

Description

technical field [0001] The invention belongs to the technical field of earthquake prediction, in particular to an earthquake prediction time-domain analysis method for extracting precursor information based on seismic wave decomposition. Background technique [0002] At present, the existing technology commonly used in the industry is as follows: accurate prediction of earthquakes includes accurate prediction of the time and location of earthquakes (hereinafter referred to as "predicting earthquakes on time and place"), because the location deviates, the time is advanced or delayed, Forecasts that are too large or too small may result in wasted time and property or increased casualties. Relevant studies have shown that early warning of earthquakes 3 seconds in advance can reduce casualties by 14%, 10 seconds in advance can reduce casualties by 39%, and 20 seconds in advance can reduce casualties by 63%. Earthquakes are caused by seismic waves produced by the sudden release ...

AI Technical Summary

Problems solved by technology

First, whether the flow of mantle material is the most likely driving force of plate movement is still controversial

Second, traditional seismology assumes that seismic waves propagate in the solid continuous crustal medium, which is inconsistent with the facts

[0010] In summary, the existing problems in the existing technology are: the dynamic mechanism of seismic wave propagation in the medium, whether seismic wave represents the seismic energy intensity, how seismic wave reflects the geological structure and dynamic changes, and the initial seismic wave and the total energy released by the earthquake These key issues are still unclear, so it is impossible to accurately extract earthquake precursor information, and it is also impossible to accurately predict earthquakes and provide information on geological structure and dynamic changes during earthquakes

Second, deep learning can use massive data to accurately predict weather forecasts like typhoons, but it cannot play a role in earthquake predictions where limited earthquake observation data is available

However, numerical solutions and simulation results are non-structural and uninterpretable, so only numerical solutions or simulations are not enough. A theoretical parameter field reflecting the earthquake dynamic mechanism is required. However, it is not easy to find this theoretical parameter field.

Images